| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Research Article |
Department of Neurobiology and Physiology,3 Northwestern University, Evanston, Illinois 60201
Howard Hughes Medical Institute,4 Chevy Chase, Maryland 20815-6789
ABSTRACT
In the female mouse, ovulation and estrous cyclicity are under both hormonal and circadian control. We have shown that mice with a mutation in the core circadian gene Clock have abnormal estrous cycles and do not have a luteinizing hormone (LH) surge on the afternoon of proestrus due to a defect at the hypothalamic level. In the present study, we tested the hypotheses that vasopressin (AVP) can act as a circadian signal to regulate the proestrous release of LH, and that this signal is deficient in the Clock mutant. We found that Avp expression in the suprachiasmatic nucleus (SCN) and AVP 1a receptor (Avpr1a) expression in the hypothalamus is reduced in Clock mutant mice compared to wild-type mice. Intracerebroventricular (i.c.v.) injection of AVP on the afternoon of proestrus is sufficient to induce LH secretion, which reaches surge levels in 50% of Clock mutant mice. The effect of AVP on the Clock mutant LH surge is mediated by AVPR1A, as co-infusion of AVP and an AVPR1A-specific antagonist prevents AVP induction of LH release, although infusion of an AVPR1A antagonist into wild-type mice failed to prevent a proestrous LH surge. These results suggest that reduced hypothalamic AVP signaling plays a role in the absence of the proestrous LH surge in Clock mutant mice. The results also support the hypothesis that AVP produced by the SCN may be a circadian signal that regulates LH release.
circadian rhythm, gonadotropin-releasing hormone, luteinizing hormone, ovulatory cycle, vasopressin
INTRODUCTION
Reproduction is dependent on a highly orchestrated cascade of events that originate from a small set of gonadotropin-releasing hormone (GnRH) neurons distributed throughout the preoptic area (POA) and septal areas of the hypothalamus. On the afternoon of proestrus, a surge of GnRH released from neuronal terminals in the mediobasal hypothalamus induces the pituitary to release luteinizing hormone (LH) and follicle stimulating hormone (FSH), which act on the ovary to induce ovulation and follicular recruitment. The preovulatory GnRH surge is primarily controlled by two types of input to GnRH neurons and the surrounding interneurons: hormonal feedback from maturing ovarian follicles and circadian output from the suprachiasmatic nucleus (SCN) [1, 2]. The absence or dysregulation of either type of input prevents GnRH surge release and results in anovulation [3].
Although proper levels of estrogen and progesterone create a permissive state for GnRH release, a timing signal from the SCN is required to induce the level of GnRH release associated with the LH surge. Evidence for the existence of this timing signal in mice is limited due to the difficulty of studying the function of the hypothalamic-pituitary-gonadal axis in this organism; however, the timing signal has been studied in both rats and hamsters, and data suggest that the same mechanism is active in mice [4, 5]. The signal from the SCN to GnRH neurons occurs once daily, as demonstrated by the fact that rats treated chronically with high levels of estradiol exhibit LH surges on multiple successive days [6]. More recently, a similar model of chronic estradiol replacement has been used to induce a daily LH surge in mice [7]. The daily signal is restricted to a relatively narrow window of time: female mice treated with pentobarbital on the afternoon, but not morning, of proestrus exhibit a 24-h delay in both the LH surge and ovulation, and a similar effect has been observed in rats. Ablation of the SCN or disruption of the neuronal projections from the SCN to the POA results in estrous acyclicity, indicating the crucial role these nuclei play in reproductive function [9–11].
It has been shown that the timing signal is most likely conveyed neuronally rather than humorally. In SCN-lesioned hamsters that have received grafts of fetal SCN tissue, circadian locomotor activity is restored but estrous cyclicity and estradiol-induced LH release are not [12, 13]. In rats, efferent SCN projections synapse directly onto GnRH neurons, and also onto estradiol-responsive interneurons adjacent to GnRH neurons [14]. Although the molecular identity of the timing signal from the SCN is currently unknown, the neuropeptides vasopressin (AVP) and vasoactive intestinal peptide (VIP) have both been implicated in temporal regulation of GnRH release [15, 16]. Vasopressin is of particular interest, as its expression in the SCN is regulated by the core circadian clock genes Clock and Arntl (also known as Bmal1), which induce Avp expression by binding to E-box elements in the AVP promoter [17–19]. Furthermore, AVP treatment of SCN-lesioned, estradiol-primed rats is sufficient to induce an LH surge [20].
We have recently shown that female Clock/Clock mutants have irregular estrous cycles and fail to have an LH surge on proestrus [21]. Because SCN expression of Avp has been shown to be significantly reduced in male Clock mutants, we hypothesized that abnormal SCN Avp expression in these animals might be the underlying cause of the Clock mutant reproductive phenotype. We examined SCN expression of Avp mRNA and hypothalamic expression of the AVP 1a receptor (Avpr1a) in ovariectomized, estradiol-treated wild-type and Clock/Clock mutant mice. We then assessed the effect of proestrous intracerebroventricular (i.c.v.) injection of an AVPR1A antagonist and/or AVP on the LH surge in wild-type and Clock mutant females in order to determine whether restoration of a putative timing signal could induce a proestrous LH surge in Clock mutants.
MATERIALS AND METHODS
Animals
Female C57BL/6J mice were maintained on a light:dark cycle of 12L:12D (lights-out = Zeitgeber Time [ZT]12), with food and water available ad libitum. Clock/Clock mice were bred in-house; due to the number of wild-type mice required for the study, wild-type controls were ordered from the Jackson Laboratory (Bar Harbor, ME) and allowed to acclimate for a minimum of 4 wk before use. All mice were between 2.5 and 5.0 mo old at the time of use. As the Clock mutation was identified on a C57BL/6J background and all Clock/Clock females were F2 progeny from the C57BL/6J stock, isogenic background was ensured. All animal and surgical procedures were conducted in accordance with the policies of Northwestern University's Animal Care and Use Committee (Evanston, IL).
In Situ Hybridization
Avp expression was measured in the SCN of wild-type (n = 35) and Clock/Clock (n = 33) mice. To control for potential effects of sex steroids on neuropeptide expression, mice were ovariectomized and estradiol-primed using a previously described protocol [22]. In our hands, this hormone priming protocol results in consistent levels of estradiol between animals and is therefore the most reliable method of inducing a coordinated "proestrous-like" state in the number of animals required for a circadian experiment. At the time of ovariectomy, mice received a single silastic implant (0.04" ID, 0.085" OD; American Scientific Products) containing 2.5 µg 17ß-estradiol (Sigma Chemical Co., St. Louis, MO) mixed with Silicone Type A Medical Adhesive (Applied Silicone Corp., Ventura, CA). Beginning at ZT20 10 days following ovariectomy, four to five mice per genotype per group were killed by cervical dislocation every 4 h for a full 24 h (ZT20–ZT20). Collections during the dark period were performed under dim red light. After sacrifice, all mice were examined for uterine ballooning to ensure the efficacy of estradiol treatment. Brains were rapidly removed, snap-frozen on dry ice, and stored at –80°C. Using a cryostat, 20-µm coronal sections through the region of the SCN were collected and thaw-mounted on positively charged slides (Superfrost Plus, VWR). In situ hybridization using radiolabeled DNA oligonucleotides was performed as previously described [23]. Briefly, a 36-bp DNA oligonucleotide probe with melting temperature of 78–82°C and GC content of 45%–55% was designed using Primer3 (http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi) against Avp (5'-TCAGGAAACAAGCGGAGAGCGTAGTGTTGAGCATCC-3') mRNA (Integrated DNA Technologies, Coralville, IA). The oligonucleotide (1 pmol/µl) was labeled using terminal deoxynucleotidyl transferase (TdT; Invitrogen, Carlsbad, CA) to add [33P]dATP to the 3' end. Unincorporated 33P was removed using Probequant G-50 Sephadex microspin columns (Amersham Pharmacia, Piscataway, NJ), and percent incorporation was determined to be 40%–60% by comparing scintillation counts of total versus incorporated probe. The probe was stored at 4°C and used within 24-h of labeling.
Brain sections were prepared for hybridization by fixation in 4% paraformaldehyde, then rinsed in 2x saline sodium citrate (SSC; 0.15 M sodium chloride and 0.015 M sodium citrate) and 0.1M TEA (pH 8.0) with acetic anhydride and dehydrated through successively more concentrated ethanol dips. The radiolabeled probe was suspended at a concentration of 2 nM in hybridization solution containing dextran sulfate, polyadenylic acid, and sheared herring sperm, applied to the brain sections, and hybridized overnight at 37°C. Post-hybridization, brain sections were washed in 55°C 1x SSC, dehydrated, dried overnight, and exposed to autoradiographic film (Kodak). Films were scanned into a computer at a resolution of 2600 dpi, and optical densities of at least three bilateral SCN per brain were measured using NIH Image (National Institutes of Health; http://rsb.info.nih.gov/nih-image/Default.html). A region of the lateral hypothalamus was used a background control.
Quantitative Real-Time Polymerase Chain Reaction
Expression of Avpr1a has previously been shown to be affected by estradiol treatment [24]. Therefore, Avpr1a expression was measured in hypothalami from mice treated with the ovariectomy-plus-estradiol protocol described above, as this high estrogen state mimics the hormonal milieu present in pro-estrus. Wild-type (n = 8) and Clock/Clock (n = 8) mice were killed at the light-dark transition six days after ovariectomy, approximating the predicted time of the circadian gating signal. Brains were rapidly removed, and a hypothalamic region extending from the POA to the mediobasal hypothalamus was dissected out. To obtain this section, the brain was placed ventral side up and a coronal cut was made at the posterior third of the olfactory tubercle. A second coronal cut was made approximately 3 mm caudally, at the rostral border of the mammillary bodies. From this section, an isosceles triangle was made with the apex just below the corpus callosum and the legs passing through the anterior commissure. The hypothalamus was snap-frozen and stored on dry ice until RNA extraction with Trizol reagent (Invitrogen). The RNA was diluted to 0.1 µg/µl prior to use with the TaqMan EZ RT-PCR real-time PCR kit (Applied Biosystems, Foster City, CA). PrimerExpress software (Applied Biosystems) and mRNA sequence data from Ensembl were used to design a probe labeled with 6-carboxyfluorescein (FAM) at its 5' terminus and 6-carboxytetramethylrhodamine (TAMRA) at its 3' terminus (Applied Biosystems) and oligonucleotide primer (Integrated DNA Technologies) sets overlapping exon junctions of the Avpr1a gene:
Avpr1a-forward, 5'-TCTTCATCGTCCAGATGTGGTC-3',
Avpr1a-reverse, 5'-CCAGTAACGCCGTGATCGT-3',
Avpr1a probe, 6FAM-CAATTTCGTTTGGACCGATTCCGAAATA-TAMRA.
RT-PCR was performed using the ABI Prism 7700 Sequence Detection System (Perkin-Elmer Applied Biosystems). Each RT-PCR reaction contained 100 ng sample mRNA, a probe and primer set for Avpr1a, and a probe and primer set for rodent Gapdh, which was used as an internal control. Probe and primer concentrations were optimized so that the target and control reactions could be performed in the same tube. Reaction conditions were as follows: Cycle 1 (1x), 50°C for 2 min; Cycle 2 (1x), 60°C for 30 min; Cycle 3 (1x), 95°C for 5 min; Cycle 4 (40x), 94°C for 20 sec, 62°C for 1 min. Expression of the target gene was determined using the delta-Ct comparative Ct (cycle of threshold detection) method to normalize target gene expression relative to Gapdh expression.
Intracerebroventricular Injections
Ovary-intact wild-type and Clock/Clock mutant mice were anesthetized with ketamine (50 mg/kg)/xylazine (10 mg/kg) and mounted in a stereotactic apparatus. A small incision (approximately 0.5-inch) was made in the skin covering the skull to obtain access to the skull surface for implantation of a guide screw cannula. A presterilized stainless steel screw (head diameter 2.5 mm, shaft diameter 1.57 mm, shaft length 2.4 mm; Plastics One, Roanoke, VA) was implanted into the skull above the anterior portion of the third ventricle and secured with cyanoacrylate glue (Plastics One). Stereotactic coordinates were, relative to bregma, AP, +0.38 mm; DM, 0 mm; VL, 0 mm [25]. The guide screw has a hole down the vertical axis through which a needle can be inserted. It functions similarly to an indwelling cannula, but eliminates the need for mounting screws or cranioplastic cement commonly required for i.c.v. cannulae [26, 27]. Mice were housed singly and allowed to recover for 4 days before beginning daily vaginal lavage to determine estrous stage. On the first nucleated smear at least 10 days after surgery, mice were briefly anesthetized with isoflurane at either ZT6 (wild-type mice only) or ZT11 (wild-type and Clock/Clock mice) and injected with the compounds described below. Wild-type mice received an injection of 2 µl artificial cerebrospinal fluid (aCSF; Fisher Scientific) or 50 ng (25 ng/µl) of a specific AVPR1A antagonist, Manning Compound ([deamino-Pen1, O-Met-Tyr2, Arg8]-vasopressin, Sigma), while Clock/Clock mice received a 2-µl injection of aCSF only, 3 ng (1.5 ng/µl) [Arg8]-vasopressin (Sigma), or a 4-µl mix containing 3 ng AVP and 50 ng Manning Compound. The dose of AVP was based on previously reported uses of intracerebral AVP in mice to affect memory [28] and in voles to effect aggression [29], and the timing of the injection was based on the previously reported critical window for the effect of AVP on LH in rats [30]. The dose of AVPR1A antagonist treatment was determined based on its previously reported effective use in rats and voles, as this compound has not been used centrally in any published reports in mice [29]. The ZT6 injection time point was chosen because our data suggest that intact wild-type mice may begin the LH surge several hours before lights-out [21], while the ZT11 injection time point was chosen based upon previous reports of the use of an AVP antagonist to reduce the LH surge in rats [31]. The injection was given using a 26 g Hamilton Syringe (Fisher Scientific), with the needle capped to a penetration depth of 6.7 mm (stereotactic coordinate DV = 5.3 mm) [25]. Following injection, mice were returned to their home cage. At ZT13.5, mice were anesthetized with isoflurane and blood was collected by cardiac puncture. The collection time point was chosen because we have previously shown that the average peak amplitude of the proestrous LH surge in wild-type mice occurs 1–2 h after lights-out [21]. Serum was collected and stored at –80°C until radioimmunoassay (RIA) for LH and FSH.
Radioimmunoassay
Luteinizing hormone RIA was performed by the Ligand Assay Core of Northwestern University (Evanston, IL). Luteinizing hormone was measured in 75 µl of serum using NIDDK-anti-rLH-S11, NIDDK-rLH-I10 for iodination, and NIDDK-rLH-RP3 for reference (National Institute of Diabetes and Digestive and Kidney Diseases [NIDDK], Rockville, MD). The minimum detectable level was 0.2 ng/ml, and the intraassay and interassay coefficients of variance were 8.85% and 9.22%, respectively.
Statistical Analysis
The expression of Avp mRNA in the SCN of wild-type and Clock mutant mice was analyzed by two-way ANOVA (time, genotype) and Least Squares (LS) Means Differences Student t-test (JMP Software, SAS Institute). Hypothalamic expression of Avpr1a mRNA was analyzed by t-test. The effect of i.c.v. AVP injection in Clock mutant mice was analyzed by one-way ANOVA (treatment); the effect of i.c.v. injection on LH release in wild-type mice was analyzed by two-way ANOVA (time, treatment). Post-hoc analysis was performed using LS Means Differences Student t-test.
RESULTS
Effect of Time and Genotype on SCN Expression of Avp
Circadian expression of Avp mRNA in the SCN was measured in ovariectomized, 17ß-estradiol-treated wild-type and Clock/Clock mutant mice. Analysis of Avp expression showed significant effects of time, genotype, and time x genotype (all P < 0.0001) (Fig. 1). There were significant differences among time points within wild-type mice: samples taken during the expected nadir of expression (ZT20 and ZT0) did not differ from one another, while samples taken at ZT4, 8, and 12 were elevated relative to those taken at the nadir. There was also an effect of time on Avp expression in Clock mutant mice, with expression at ZT8 and ZT12 significantly elevated relative to expression at all other time points. At all time points, Avp mRNA expression in wild-type SCN was significantly elevated compared with expression in Clock mutants (mean difference = 87 optical density [OD]). It is important to note that, as shown by Jin et al. [32], the effect of the Clock mutation on Avp expression is specific to the SCN: expression of Avp in the supraoptic nucleus (SON) is normal and non-rhythmic in Clock mutants.
|
Effect of Genotype on Hypothalamic Avpr1a Expression
Hypothalamic expression of Avpr1a was measured in ovariectomized, estradiol-primed wild-type and Clock/Clock mice. The relative abundance of Avpr1a mRNA was significantly reduced (P = 0.026) in Clock mutants compared with wild-type mice (Fig. 2).
|
Effect of AVP and/or AVPR1A Antagonists on the Proestrous LH Surge
Clock/Clock mutants received an injection of aCSF (n = 19), AVP (n = 18), or a mix of AVP + AVPR1A antagonist (n = 13) into the anterior third ventricle at ZT11 on the day of a nucleated vaginal smear, and serum LH levels were measured at ZT13.5. There was a significant effect of treatment (P = 0.001) on LH levels: AVP treatment resulted in a significant elevation of LH over both the aCSF-treated group (P < 0.001) and the AVP + AVPR1A antagonist group (P = 0.003) (Fig. 3). Mean LH levels per group were: aCSF, 0.92 ng/ml (standard error [SE] ± 0.24); AVP, 7.38 ng/ml (SE ± 1.75); AVP + AVPR1A antagonist, 0.99 ng/ml (SE ± 0.52).
|
Wild-type mice received an injection of aCSF or a specific AVPR1A antagonist into the anterior third ventricle at either ZT6 (aCSF, n = 9; AVPR1A antagonist, n = 10) or ZT11 (aCSF, n = 12; AVPR1A antagonist, n = 13) on the day of a nucleated vaginal smear. Serum LH levels were measured at ZT13.5. Analysis of time and treatment effect showed that AVPR1A antagonist treatment at ZT6 (Fig. 4A) or ZT11 (Fig. 4B) had no effect on the proestrous LH surge. There was a decrease in proestrous LH in the ZT11 AVPR1A-antagonist group, but this effect was not significant (P = 0.29). Mean LH values per group were as follows: ZT6 aCSF, 3.87 ng/ml (SE ± 1.59) and ZT6 AVPR1A antagonist, 5.65 ng/ml (SE ± 2.3); ZT11 aCSF, 9.65 ng/ml (SE ± 2.4) and ZT11 AVPR1A antagonist, 5.37 ng/ml (SE ± 1.5).
|
In order to verify the stereotactic coordinates and the distribution of injected material, seven wild-type and Clock/Clock mice received an i.c.v. injection of 5 µl 0.4% Trypan Blue solution (Sigma) at ZT11, and were killed 2.5 h later. Brains were snap-frozen on dry ice, sectioned into 40-µm sections by cryostat, counterstained with Nuclear Fast Red (Vector Laboratories, Burlingame, CA), and examined by microscopy. Injections were found to be in the anterior third ventricle at the level of the organum vasculosum of the lamina terminalis/medial POA (mPOA) (Fig. 4C). For all experimental mice, Trypan Blue was injected postmortem through the guide screw, and the brain was rapidly dissected to verify the site of injection.
DISCUSSION
Vasopressin has been posited as a candidate molecule for the SCN-derived daily timing signal that regulates the proestrous GnRH surge. In this study, we have shown that 1) Avp expression in the SCN and Avpr1a expression in the hypothalamus are reduced in female Clock/Clock mutant mice, 2) a single injection of AVP into the region of the medial POA on the afternoon of proestrus is sufficient to induce the LH surge in half of AVP-treated Clock mutants, while none of the control mice had an LH surge, and 3) the effect of AVP on induction of the proestrous LH surge in Clock mutants is mediated by AVPR1A. In addition to demonstrating that AVP may act as a circadian timing signal for LH release in mice, this is the first report, to our knowledge, of the pharmacological rescue of a physiological phenotype in a mouse carrying a mutation in a core clock gene.
We have previously shown that Clock mutant female mice exhibit pleiotropic reproductive defects, including irregular estrous cyclicity and failure to have a coordinated LH surge on proestrus. We have also shown that the lack of an LH surge is due to a defect at the hypothalamic level: although hypothalamic GnRH peptide content, pituitary response to GnRH, and circulating levels of estradiol and progesterone on proestrus are normal in Clock/Clock females, these mutant mice fail to respond to estradiol priming with an LH surge [21]. The lack of an LH surge in intact proestrous or ovariectomized, estradiol-primed Clock mutants is therefore due to either an inadequate daily timing signal, an inability of GnRH neurons or GnRH-modulating interneurons to respond to a circadian timing signal, a failure of estradiol to induce necessary changes in neuronal activity at the hypothalamic level, or a combination of these factors. Given the role that AVP plays in LH release and the evidence that Avp expression is dramatically reduced in Clock/Clock mice, we hypothesized that the low level and arrhythmicity of Avp expression might be the cause of the proestrous phenotype observed in Clock mutants. Our results support this hypothesis: in 50% of the Clock mutants, i.c.v. injection of AVP in the late afternoon of proestrus induced an appropriately timed LH surge, whereas none of the Clock mutants receiving vehicle alone had an LH surge. The effect we observed of AVP on LH release was mediated by AVPR1A, as co-injection of AVP with an AVPR1A antagonist prevented AVP from inducing an LH surge in 85% of treated Clock mutants. Future experiments aimed at assessing the effect on LH release of AVP injections at different times throughout the day in estradiol-primed mutants may further support the role of AVP as a circadian gating signal for the reproductive axis.
Although only half the AVP-injected Clock mutants exhibited an LH surge, this percentage is similar to the percent of vehicle-treated wild-type mice that had an LH surge. It is possible that, had we sampled more frequently, we would have observed elevated LH in more mice. However, we have previously performed a detailed analysis of the timing of the proestrous LH surge by measuring serum LH in wild-type and Clock/Clock mice sampled hourly between ZT5 and ZT21 on the day of a nucleated (proestrus) vaginal smear [21]. In addition to determining that none of the Clock mutant mice had an LH surge at any time in proestrus, we found that only 50% of wild-type mice had an LH surge, and that ZT13 and ZT14 represented the only time when the surges of all individuals overlapped. Because surgeries to implant both an indwelling atrial cannula and an i.c.v. cannula would be highly stressful to an animal as small and reproductively sensitive as a mouse, we chose ZT13.5 as the most appropriate time to perform a single blood collection.
We did not test the critical period for AVP injection, but previous studies in rats have suggested that the effective period for AVP stimulation of LH release occurs in the late afternoon of proestrus, shortly before the expected onset of the LH surge. Palm et al. [30] showed that AVP infusion from ZT7.5–ZT12.5, but not ZT0–ZT5, increased the amplitude of the LH surge in SCN-intact, estradiol-treated rats, while Funabashi et al. [31] showed that a single injection of a nonspecific vasopressin receptor antagonist at ZT8, approximately 2 h prior to the expected onset of the LH surge, significantly reduced peak LH amplitude. Studies using pentobarbital to delay the LH surge in mice, and our data on the timing of LH surge in intact wild-type mice, suggest that the onset of the LH surge is slightly phase-delayed in mice relative to rats [4, 21]. Our results support the hypothesis that the timing signal occurs shortly before onset of the LH surge [8]. However, we were unable to replicate the inhibitory effects of the vasopressin receptor antagonist on LH release in wild-type mice. This may be due to the fact that the receptor antagonist we used was specific for AVPR1A, while the antagonist used by Funabashi et al. activates AVPR1A, AVPR1B, and oxytocin receptors [33]. Despite our inability to suppress LH release in wild-type mice with a highly specific AVPR1A antagonist, our data support the hypothesis that AVPR1A plays a significant role in mediating the release of LH in Clock mutant mice: co-injection of a specific AVPR1A antagonist blocked the effects of AVP on LH release in Clock/Clock mice. Therefore, we suggest that local i.c.v. injection of AVP was sufficient to stimulate a subset of GnRH neurons, leading to release of LH in Clock mutants, but that the AVPR1A antagonist injected i.c.v. in wild-type mice failed to overcome the activational effect of endogenous AVP release, possibly because GnRH neurons outside the diffusion range of the antagonist are sufficient to induce an LH surge. It is also possible that multiple neuropeptides and/or neurotransmitters that may be deficient in Clock mutants normally contribute to the LH surge in wild-type mice; in this case, inhibiting AVPR1A function might not be sufficient to prevent the wild-type LH surge.
The molecular identity of the daily timing signal for the reproductive axis has been controversial: both Avp and Vip are rhythmically expressed and present in SCN neurons efferent to the mPOA, and, in rats, inhibition of either AVP or VIP signaling results in reduction of the amplitude of an estradiol-induced LH surge [31, 34, 35]. However, AVP alone is capable of inducing an LH surge in SCN-lesioned rats, and if the rhythms of AVP and VIP are phase-dissociated in SCN-POA co-cultures, AVP secretion occurs in phase with GnRH release, whereas VIP secretion does not [20, 36]. Furthermore, previous studies have shown that transcription of Avp in the SCN, but not other brain regions, is under control of the CLOCK protein, and that Avp, but not Vip, expression is reduced in the SCN of Clock/Clock mutant male mice [36, 37]. The present experiments do not rule out the possibility that VIP contributes to the regulation of either the timing or the amplitude of the LH surge, as has previously been proposed [34]; however, the results do provide strong evidence for a role of AVP in the generation of the LH surge in a circadian mutant.
In addition to low SCN Avp expression, Clock/Clock mutants exhibit reduced expression of the vasopressin 1a receptor. Avpr1a, which encodes the major AVP receptor subtype in the hypothalamus [38], and is expressed ion estradiol-sensitive neurons in the POA [24], may be key to the reproductive axis response to circadian signals. Expression of Avpr1a is upregulated by estrogen in the rat hypothalamus; neurons that respond to both AVP and estradiol and innervate GnRH neurons may represent a physical intersection for the hormonal and circadian signals that induce the GnRH surge [24, 36, 39]. We suggest that, even if the Clock mutant SCN is capable of transmitting a low amplitude AVP signal to GnRH-innervating interneurons, these neurons may lack the sensitivity necessary to respond appropriately.
In addition to implicating deficient AVP signaling in the failure of Clock mutants to have a proestrous LH surge, our data allow us to hypothesize a model for the interaction between hormonal signals and at least one circadian signal that regulates the LH surge in wild-type mice. Each afternoon, neurons that project from the SCN to interneurons that innervate GnRH neurons release a high level of AVP, as circadian production of AVP in the SCN reaches a peak. In the absence of positive feedback from estradiol, AVP has little or no effect on GnRH release. However, if estradiol is present, it upregulates the expression of Avpr1a in hypothalamic neurons, making these neurons more sensitive to the stimulatory AVP signal from the SCN. In Clock mutant mice, the timing signal is lacking. The current data support a role for AVP as this timing signal, as Avp expression in the SCN of Clock/Clock females is low, with peak Avp expression equivalent to the Avp expression nadir in wild-type mice. Clock mutants are at an additional disadvantage because they express very low levels of Avpr1a. However, despite low hypothalamic Avpr1a expression, AVP treatment alone is sufficient to induce a proestrous LH surge in Clock mutants. Therefore, it may be that there is redundancy in the circadian signals that control the GnRH surge; if all components are abnormal, the surge will not occur, but if one component is restored, the barrier to surge induction is overcome.
ACKNOWLEDGMENTS
We thank Brigitte Mann for performing radioimmunoassays.
FOOTNOTES
2 Correspondence and current address: Brooke H. Miller, TSRI-Scripps Florida, Jupiter, FL 33458. FAX: 561 799 8957; bmiller{at}scripps.edu 
1 Supported by the Howard Hughes Medical Institute to J.S.T. and fellowships from the National Science Foundation and National Institute of Neurological Disorders and Stroke to B.H.M.
Received: 29 March 2006.
First decision: 25 April 2006.
Accepted: 19 July 2006.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  C. A. Christian and S. M. Moenter Vasoactive Intestinal Polypeptide Can Excite Gonadotropin-Releasing Hormone Neurons in a Manner Dependent on Estradiol and Gated by Time of Day Endocrinology, June 1, 2008; 149(6): 3130 - 3136. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. SILVER and J. LESAUTER Circadian and Homeostatic Factors in Arousal Ann. N.Y. Acad. Sci., May 1, 2008; 1129(1): 263 - 274. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. O. Poletini, D. T. McKee, J. E. Kennett, J. Doster, and M. E. Freeman Knockdown of clock genes in the suprachiasmatic nucleus blocks prolactin surges and alters FRA expression in the locus coeruleus of female rats Am J Physiol Endocrinol Metab, November 1, 2007; 293(5): E1325 - E1334. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
